| { | |
| "_name_or_path": "/ceph/hdd/staff/charpent/.cache/modelsk31314q8h8ug76yp", | |
| "architectures": [ | |
| "MiniCPMForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_minicpm.MiniCPMConfig", | |
| "AutoModel": "openbmb/MiniCPM-2B-dpo-bf16--modeling_minicpm.MiniCPMModel", | |
| "AutoModelForCausalLM": "modeling_minicpm.MiniCPMForCausalLM", | |
| "AutoModelForSeq2SeqLM": "openbmb/MiniCPM-2B-dpo-bf16--modeling_minicpm.MiniCPMForCausalLM", | |
| "AutoModelForSequenceClassification": "openbmb/MiniCPM-2B-dpo-bf16--modeling_minicpm.MiniCPMForSequenceClassification" | |
| }, | |
| "bos_token_id": 1, | |
| "dim_model_base": 256, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 2304, | |
| "initializer_range": 0.1, | |
| "intermediate_size": 5760, | |
| "max_position_embeddings": 4096, | |
| "model_type": "minicpm", | |
| "num_attention_heads": 36, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 36, | |
| "pretraining_tp": 1, | |
| "quantization_config": { | |
| "_load_in_4bit": true, | |
| "_load_in_8bit": false, | |
| "bnb_4bit_compute_dtype": "bfloat16", | |
| "bnb_4bit_quant_storage": "uint8", | |
| "bnb_4bit_quant_type": "fp4", | |
| "bnb_4bit_use_double_quant": false, | |
| "llm_int8_enable_fp32_cpu_offload": false, | |
| "llm_int8_has_fp16_weight": false, | |
| "llm_int8_skip_modules": [ | |
| "lm_head" | |
| ], | |
| "llm_int8_threshold": 6.0, | |
| "load_in_4bit": true, | |
| "load_in_8bit": false, | |
| "quant_method": "bitsandbytes" | |
| }, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "scale_depth": 1.4, | |
| "scale_emb": 12, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.40.0", | |
| "use_cache": true, | |
| "vocab_size": 122753 | |
| } | |